Code from BasicOptimizer.scala:75 executed in 725.41 seconds (4.482 gc):
val lineSearchInstance: LineSearchStrategy = lineSearchFactory
IterativeTrainer.wrap(trainable)
.setOrientation(orientation())
.setMonitor(new TrainingMonitor() {
override def clear(): Unit = trainingMonitor.clear()
override def log(msg: String): Unit = trainingMonitor.log(msg)
override def onStepFail(currentPoint: Step): Boolean = {
BasicOptimizer.this.onStepFail(trainable, currentPoint)
}
override def onStepComplete(currentPoint: Step): Unit = {
if (0 < logEvery && (0 == currentPoint.iteration % logEvery || currentPoint.iteration < logEvery)) {
val image = currentImage
timelineAnimation += image
val caption = "Iteration " + currentPoint.iteration
out.p(caption + "\n" + out.jpg(image, caption))
}
BasicOptimizer.this.onStepComplete(trainable, currentPoint)
trainingMonitor.onStepComplete(currentPoint)
super.onStepComplete(currentPoint)
}
})
.setTimeout(trainingMinutes, TimeUnit.MINUTES)
.setMaxIterations(trainingIterations)
.setLineSearchFactory((_: CharSequence) => lineSearchInstance)
.setTerminateThreshold(java.lang.Double.NEGATIVE_INFINITY)
.runAndFree
.asInstanceOf[lang.Double]
Reset training subject: 944673901679800
Reset training subject: 944719451469900
Adding measurement fff51b8 to history. Total: 0
LBFGS Accumulation History: 1 points
Constructing line search parameters: GD+Trust
th(0)=4.159089580178261;dx=-1.96922825279083E-4
New Minimum: 4.159089580178261 > 4.158759728074074
WOLFE (weak): th(2.154434690031884)=4.158759728074074; dx=-1.861225877254693E-4 evalInputDelta=3.298521041870117E-4
New Minimum: 4.158759728074074 > 4.158435121178627
WOLFE (weak): th(4.308869380063768)=4.158435121178627; dx=-1.860552062075927E-4 evalInputDelta=6.544589996337891E-4
New Minimum: 4.158435121178627 > 4.15713857114315
WOLFE (weak): th(12.926608140191302)=4.15713857114315; dx=-1.8540109101429421E-4 evalInputDelta=0.0019510090351104736
New Minimum: 4.15713857114315 > 4.151357010006905
WOLFE (weak): th(51.70643256076521)=4.151357010006905; dx=-1.831848923848072E-4 evalInputDelta=0.007732570171356201
New Minimum: 4.151357010006905 > 4.121324852108955
END: th(258.53216280382605)=4.121324852108955; dx=-1.762137079805682E-4 evalInputDelta=0.03776472806930542
Fitness changed from 4.159089580178261 to 4.121324852108955
Iteration 1 complete. Error: 4.121324852108955 Total: 408.2108; Orientation: 0.0620; Line Search: 271.9963
<a id="p-3"></a>Iteration 1
<a id="p-2"></a>
Adding measurement 2e5b1aa8 to history. Total: 1
LBFGS Accumulation History: 2 points
th(0)=4.121324852108955;dx=-1.6633722090772217E-4
New Minimum: 4.121324852108955 > 4.050034560263157
WOLFE (weak): th(556.9906600335335)=4.050034560263157; dx=-1.5055543524932554E-4 evalInputDelta=0.07129029184579849
New Minimum: 4.050034560263157 > 3.9865404814481735
END: th(1113.981320067067)=3.9865404814481735; dx=-1.3354854998577634E-4 evalInputDelta=0.13478437066078186
Fitness changed from 4.121324852108955 to 3.9865404814481735
Iteration 2 complete. Error: 3.9865404814481735 Total: 181.7218; Orientation: 0.0652; Line Search: 136.3358
<a id="p-5"></a>Iteration 2
<a id="p-4"></a>
Adding measurement 796fbd4e to history. Total: 2
LBFGS Accumulation History: 3 points
th(0)=3.9865404814481735;dx=-1.1544393058474284E-4
New Minimum: 3.9865404814481735 > 3.8024019300937653
END: th(2400.0)=3.8024019300937653; dx=-8.254475501232279E-5 evalInputDelta=0.18413855135440826
Fitness changed from 3.9865404814481735 to 3.8024019300937653
Iteration 3 complete. Error: 3.8024019300937653 Total: 135.4121; Orientation: 0.1045; Line Search: 90.4324
<a id="p-7"></a>Iteration 3
<a id="p-6"></a>
Final threshold in iteration 3: 3.8024019300937653 (> -Infinity) after 725.405s (< 720.000s)
3.8024019300937653